CN102240212A - Method and apparatus for measuring pneumothorax - Google Patents

Method and apparatus for measuring pneumothorax Download PDF

Info

Publication number
CN102240212A
CN102240212A CN2010101803339A CN201010180333A CN102240212A CN 102240212 A CN102240212 A CN 102240212A CN 2010101803339 A CN2010101803339 A CN 2010101803339A CN 201010180333 A CN201010180333 A CN 201010180333A CN 102240212 A CN102240212 A CN 102240212A
Authority
CN
China
Prior art keywords
thorax
image
pneumothorax
right sides
obtains
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN2010101803339A
Other languages
Chinese (zh)
Other versions
CN102240212B (en
Inventor
赵静
焦洁
史轶伦
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
GE Medical Systems Global Technology Co LLC
Original Assignee
GE Medical Systems Global Technology Co LLC
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by GE Medical Systems Global Technology Co LLC filed Critical GE Medical Systems Global Technology Co LLC
Priority to CN201010180333.9A priority Critical patent/CN102240212B/en
Publication of CN102240212A publication Critical patent/CN102240212A/en
Application granted granted Critical
Publication of CN102240212B publication Critical patent/CN102240212B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Abstract

The invention relates to a method and an apparatus for measuring pneumothorax. An X-ray CT (computed tomography) device for measuring pneumothorax comprises: a scanning platform, a data collection unit, an image reconstruction unit, a user operation interface for user operation, a central control unit and a pneumothorax measurement unit, wherein the pneumothorax measurement unit is used for the pneumothorax measurement of the chest image reconstructed by the image reconstruction unit and comprises a left-and-right thorax splitting unit, a pneumothorax splitting unit and a calculation unit; the left-and-right thorax splitting unit is used for splitting left and right thoraxes of the original chest image under the control of the central control unit to obtain a chest image with left and right thoraxes separated; the pneumothorax splitting unit is used for splitting the original chest image to obtain an image only with pneumothorax; and the calculation unit is used for calculating the areas of the left and right thoraxes according to the chest image with left and right thoraxes separated obtained through the splitting of the left-and-right thorax splitting unit, calculating the area of the pneumothorax according to the image only with pneumothorax obtained by the splitting of the pneumothorax splitting unit and calculating the proportion of the pneumothorax in the left and right thoraxes according to the following formula.

Description

Measure the method and apparatus of pneumothorax
[technical field]
The present invention relates to field of medical technology, particularly method and apparatus about pneumothorax being measured based on computerized tomography X ray (CT) image.
[background technology]
At first there is the protection of one deck visceral pleura the lung outside of human body, and the protection of one deck parietal pleura is arranged in the outside of visceral pleura, and the parietal pleura outside is exactly a thorax.Be to coincide to such an extent that finely do mutual motion between visceral pleura and the parietal pleura under the situation of normal lung.If but the parietal pleura of lung has problem, can produce leakage and make between air admission visceral pleura and the parietal pleura, thereby form said pneumothorax.Pneumothorax (PTX) is a kind of disease, and it can oppress the normal expiration of lung and air-breathing, when this pneumothorax reaches certain ratio in thorax, can constitute life danger.
At present the doctor mainly contains two kinds of methods based on graphical analysis to the measuring and calculating of pneumothorax:
1) adopts the manual method of cutting apart manual calculation.The doctor is the one-sided thorax area of the every layer of pneumothorax area of manual calculations and the left and right sides on the CT chest image, again accumulation calculating pneumothorax content.This method need spend great amount of manpower and time.
2) based on the X-ray sheet of coronalplane, as No. 5668888 patent of United States Patent (USP).The corresponding relation of the physical dimension of pneumothorax, geometry and pneumothorax volume content is derived empirical equation on the coronalplane.This technology still needs the pneumothorax on the manual analyzing coronalplane on the one hand, and empirical equation does not have universality widely on the other hand.
Some researchs based on the automatic division method of CT scan picture are also arranged in addition.But these researchs all not systematically proposition method will cut apart automatically and be integrated in the CT device, form the complete pneumothorax measuring system of a cover.Simultaneously, in dividing method, do not consider cutting apart automatically and the automatic removal partly of non-pneumothorax such as trachea, bronchus of left and right sides thorax; Simultaneously, on manual methods, only provide manual crisperding function, do not allow after the user's modification, carry out once more and cut apart automatically, cause the result of calculation inaccuracy, may mislead the doctor and do the diagnosis that makes mistake and there are to propose some parameters of cutting apart automatically.
[summary of the invention]
Purpose of the present invention is intended to address the above problem, and proposes a kind of device of pneumothorax being measured based on the CT image, and its measurement result degree of accuracy is than higher.
Another object of the present invention is intended to address the above problem, and proposes a kind of method of pneumothorax being measured based on the CT image, and its measurement result degree of accuracy is than higher.
The X ray CT device of measurement pneumothorax provided by the invention comprises: scan table, and it carries out ray scanning to person under inspection's chest; Data collection module, its scan-data to scan table output is collected, analog digital conversion; Image reconstruction unit, its scan-data that sends according to data collection module carries out image reconstruction, and the image data storage that institute's reconstruct is obtained is to a memory element; User interface for user's operation; And central control unit, it is connected to the outfan of described image reconstruction unit, and gated sweep platform, data collection module and user interface; Central control unit receives the operational order that user interface sends, and the image that image reconstruction unit is sent is handled, and shown in user interface; Pneumothorax measuring unit, its chest image after to image reconstruction unit reconstruct carry out pneumothorax and measure.
Wherein, described pneumothorax measuring unit comprises left and right sides thorax cutting unit, pneumothorax cutting unit, computing unit; Wherein, left and right sides thorax cutting unit carries out left and right sides thorax to original chest image and cuts apart under the control of central control unit, obtains the isolating thorax image of left and right sides thorax; The position of the left and right sides thorax image that described pneumothorax cutting unit splits according to left and right sides thorax cutting unit extracts original thorax image from original chest image, and this original thorax image is cut apart, and obtains the pneumothorax image; Computing unit is cut apart the area of the isolating thorax image calculation of the left and right sides thorax left and right sides thorax of acquisition according to left and right sides thorax cutting unit; Cut apart the pneumothorax image calculation pneumothorax area of acquisition according to the pneumothorax cutting unit, and calculate the ratio of pneumothorax in the thorax of the left and right sides according to following formula,
PTXpercentage = Σ i = 1 N PTXarea Σ i = 1 N HemithoraxArea
Wherein, Hemithorax Area represents the left thorax in each cross section or the area of right thorax, comprises left lung or right lung tissue that pneumothorax that each cross section splits adds each cross section;
PTX area represents the area of the pneumothorax that each cross section splits;
The quantity in ' N ' expression cross section.
Wherein, described left and right sides thorax cutting unit also comprises thorax acquiring unit, judging unit and left and right sides thorax separative element; Wherein the thorax acquiring unit carries out gray level threshold segmentation to original chest image, obtains threshold value and cuts apart the original thorax image in back, and this image is the two-value gray level image; Judging unit judges the thorax image that the thorax acquiring unit obtains judge whether left and right sides thorax separates, and instruction is sent to central control unit, and central control unit is according to instruction, the work of control left and right sides thorax separative element; If it is isolating that judged result needs, central control unit control left and right sides thorax separative element is cut apart the thorax image, separates left and right sides breast exterior feature; If judged result is not need isolatingly, then controls left and right sides thorax separative element and do not work; The needs that left and right sides thorax separative element sends the thorax acquiring unit carry out isolating thorax image and carry out left and right sides thorax separation.
Wherein, original chest image is carried out gray level threshold segmentation is to adopt maximum variance between clusters to the thorax acquiring unit; The thorax acquiring unit removes the part of non-thorax from the two-value gray level image that obtains, and finds maximum UNICOM zone and its border, removes all targets outside this border, obtains the thorax zone; And the thorax zone filled, obtain complete thorax image.
Wherein, judging unit judges according to the number of connected region with according to the area of connected region whether left and right sides thorax connects together.
Wherein, the needs that left and right sides thorax separative element sends the thorax acquiring unit carry out isolating thorax image and carry out distance conversion acquisition distance map, and carry out left and right sides thorax based on this distance map and separate, obtain isolating left and right sides thorax image, and from the image of isolating left and right sides thorax, shift out trachea, bronchus.
Wherein, left and right sides thorax separative element carries out left and right sides thorax by following step to be separated, and obtains the isolating thorax contour images of left and right sides thorax:
Step 231 is obtained the two-value gray level image of complete thorax image;
Step 232 is called the expansion function earlier to the gray level image in the step 231, calls the corrosion function again and realizes closed operation, reaches level and smooth purpose;
Step 233 is carried out the conversion of gray scale and distance to the image in the step 232, produces a range image; The gray value of the every bit in the distance map is this point minimum range to background in the two-value gray-scale map;
Step 234 is provided with a high gray threshold, cuts apart range image, judges whether left and right sides thorax separates, if do not separate, resets a higher gray threshold, cuts apart range image, separates up to left and right sides thorax, obtains the distance map of isolating left and right sides thorax;
Step 235 reduces gray threshold gradually, and with the distance map of the isolating left and right sides thorax of acquisition in the segmentation procedure 234, thorax links up up to the left and right sides, and preserves result of this time cutting apart and the result of cutting apart last time;
Step 236 deducts the result of cutting apart last time by this result of cutting apart, obtains difference results;
Step 237 in the last time segmentation result, from top to bottom, from left to right travels through each and belongs to point in the difference results, detects its adjacent area;
Step 238 is cut apart the last time and is found out those points in the previous segmentation result, and its adjacent area comprises the point that is positioned at left thorax and right thorax simultaneously, these tie-points as left and right sides thorax that record is found;
Step 239 removes the junction point and these junction points that makes progress in front and back that are found in the thorax gray level image, separate up to left and right sides thorax, obtains the isolating image of left and right sides thorax.
Wherein, confirm trachea, bronchus according to region area and circularity, for area less than the gross area 2 ‰, circularity is considered to trachea or bronchus greater than 0.9 and is removed; Circularity can be calculated by following mode:
RoundMetric = π × EquivDiameter Perimeter
Wherein, EquivDiameter is an equivalent diameter, and it is the circular cross-section diameter that converses according to area-of-interest area equal principle; Perimeter is meant the girth of area-of-interest.
Wherein, the pneumothorax cutting unit carries out threshold value to the original thorax image that extracts to be cut apart, and obtains to have only the image of pneumothorax.Original thorax image is carried out CT value threshold value cut apart, obtain the pneumothorax image.
Wherein, computing unit also comprises taxon, classification about it carries out the pneumothorax result according to geometrical property.
Wherein, described user interface comprises display interface and operation interface, display interface is used to show original chest image, the image in each stage comprised two-value gray level image, thorax image, thorax distance map, isolating left and right sides thorax image during left and right sides thorax separated, and the image in the pneumothorax cutting procedure.
Wherein, operation interface comprises that automatic pneumothorax computing function button, thorax are cut apart button, pneumothorax is cut apart button, pneumothorax computed push-buttom; Wherein cut apart button, pneumothorax and cut apart the corresponding respectively input frame that is provided with gray scale, area, circular arc degree under the button at thorax; The user can select automatic pneumothorax computing function button that pneumothorax is calculated automatically, measures; Operation thorax that also can be manual is cut apart button, pneumothorax is cut apart button, pneumothorax computed push-buttom, revises the interior value of input frame of gray scale, area, circular arc degree, carries out manual operation, calculates pneumothorax.
The method of measurement pneumothorax provided by the invention comprises the steps:
Step 1 is carried out CT scan to chest, obtains all original chest images;
Step 2 is cut apart the thoracic cavity each original chest image, obtains the complete isolating image of left and right sides thorax;
Step 3 is cut apart pneumothorax each original chest image, isolates pneumothorax, obtains the image of pneumothorax;
Step 4 is calculated the pneumothorax ratio in every one-sided thorax.
Wherein, obtain the isolating image of left and right sides thorax in the step 2 and comprise following step:
Step 21 is carried out automatic gray level threshold segmentation based on maximum variance between clusters to original chest image, obtains threshold value and cuts apart the original thorax figure in back, and this original thorax figure is the two-value gray level image;
Step 22 removes the part of non-thorax among the original thorax figure of gray level threshold segmentation that obtains from step 21, obtain complete thorax image;
By the thorax image is judged, whether need to carry out left and right sides thorax and separate, if carry out step 23; If obtain that left and right sides thorax has separated in the image, do not need to separate, then carry out step 24;
Step 23 is carried out the distance conversion to the thorax image that obtains in the step 22 and is obtained the thorax distance map, and carries out left and right sides thorax based on this thorax distance map and separate, and obtains isolating left and right sides thorax image;
Step 24 shifts out trachea, bronchus from the image of isolating left and right sides thorax.
Wherein, in the image that step 21 obtains, find maximum UNICOM zone and its border, remove all targets outside this border, obtain thorax, and chest is filled, obtain complete thorax image, fill very little, highdensity blood vessel in the pulmonary in the thorax, obtain complete thorax image.
Wherein, step 23 is carried out left and right sides thorax separation by following step:
Step 231, the two-value gray level image of the complete thorax image in the obtaining step 22;
Step 232 is called the expansion function earlier to the gray level image in the step 231, calls the corrosion function again and realizes closed operation, reaches level and smooth purpose;
Step 233 is carried out the conversion of gray scale and distance to the image in the step 232, produces a range image; The gray value of the every bit in the distance map is this point minimum range to background in the two-value gray-scale map;
Step 234 is provided with a high gray threshold, cuts apart range image, judges whether left and right sides thorax separates, if do not separate, resets a higher gray threshold, cuts apart range image, separates up to left and right sides thorax, obtains the distance map of isolating left and right sides thorax;
Step 235 reduces gray threshold gradually, and with the distance map of the isolating left and right sides thorax of acquisition in the segmentation procedure 234, thorax links up up to the left and right sides, and preserves result of this time cutting apart and the result of cutting apart last time;
Step 236 deducts the result of cutting apart last time by this result of cutting apart, obtains difference results;
Step 237 in the last time segmentation result, from top to bottom, from left to right travels through each and belongs to point in the difference results, detects its adjacent area;
Step 238 is cut apart the last time and is found out those points in the previous segmentation result, and its adjacent area comprises the point that is positioned at left thorax and right thorax simultaneously, these tie-points as left and right sides thorax that record is found;
Step 239 removes the junction point and these junction points that makes progress in front and back that are found in the thorax gray level image, separate up to left and right sides thorax, obtains the isolating image of left and right sides thorax.
Wherein, judge whether it is trachea or bronchus by region area and circularity, for those areas less than the gross area 2 ‰, circularity is considered to trachea or bronchus greater than 0.9 and should removes; Circularity is calculated by following mode:
RoundMetric = π × EquivDiameter Perimeter
Wherein, EquivDiameter is an equivalent diameter, and it is the circular cross-section diameter that converses according to area-of-interest area equal principle; Perimeter is meant the girth of area-of-interest.
Wherein, in the step 3, cut apart pneumothorax by following mode:
Step 31, according to the position of the left and right sides thorax image that obtains in the step 2 from the original thorax image of extracting of original chest image correspondence;
Step 32 is carried out threshold value to the original thorax image that extracts in the step 31 and is cut apart, and obtains to have only the image of pneumothorax.
Wherein, adopt the CT value to cut apart, obtain to have only the image of pneumothorax original thorax image.
Wherein,, calculate the ratio of pneumothorax in thorax, comprise following step in the step 4:
Step 41, classification about according to geometrical property the pneumothorax result of cutting apart being carried out;
Step 42 is according to the ratio of pneumothorax in the following formula calculating left and right sides thorax;
PTXpercentage = Σ i = 1 N PTXarea Σ i = 1 N HemithoraxArea
Wherein, Hemithorax Area represents the left thorax in each cross section or the area of right thorax, comprises left lung or right lung tissue that pneumothorax that each cross section splits adds each cross section; PTX area represents the area of the pneumothorax that each cross section splits;
The quantity in ' N ' expression cross section.
Left and right sides thorax is cut apart automatically, pneumothorax is cut apart by original chest image is carried out in the present invention, obtain isolating left and right sides thorax image and independent pneumothorax image, by about classification, calculate the ratio of the pneumothorax in the thorax of the left and right sides, the computational accuracy height has made things convenient for doctor's operation simultaneously.In addition, by the trachea in the thorax, bronchus are removed, further improved this method and device certainty of measurement among the present invention to pneumothorax.
[description of drawings]
Fig. 1 is the flow chart of measurement pneumothorax provided by the invention;
Fig. 2 is the further refinement flow chart that step 2 arrives step 4 among Fig. 1;
Fig. 3 a is the result schematic diagram of cutting apart each stage in the thorax process of the left and right sides in the step 2 to Fig. 3 f;
Fig. 4 is for obtaining the schematic flow sheet that separates left and right sides thorax among the present invention;
Fig. 5 a is cut apart the result schematic diagram in each stage in the pneumothorax process for the present invention to Fig. 5 b;
Fig. 6 measures the CT apparatus function module diagram of pneumothorax for the present invention;
Fig. 7 measures the UI interface sketch map of the CT device of pneumothorax for the present invention.
[specific embodiment]
Below, describe example of the present invention in detail with reference to accompanying drawing.The invention is not restricted to example.
See also shown in Figure 1ly, the method journey figure for the present invention measures pneumothorax comprises the steps:
Step 1 is carried out CT scan to chest, obtains all original chest images relevant with pulmonary, if shown in the 3a;
Step 2 is cut apart the thoracic cavity each original chest image, obtains the complete isolating image of left and right sides thorax, shown in Fig. 3 f;
Step 3 is cut apart pneumothorax each original chest image, isolates pneumothorax, obtains the image of pneumothorax, shown in Fig. 5 b;
Step 4 is calculated pneumothorax ratio in every one-sided thorax.
Please also refer to Fig. 2, Fig. 2 is the further refinement flow chart that step 2 arrives step 4 among Fig. 1.Wherein, obtain the isolating image of left and right sides thorax in the step 2 and comprise following step:
Step 21, based on maximum variance between clusters (Ostu method) original chest image (shown in Fig. 3 a) is carried out automatic gray level threshold segmentation, obtain threshold value and cut apart the original thorax figure in back, shown in Fig. 3 b, this original thorax figure is the two-value gray level image, this gradation of image transition is bigger, has only B﹠W basically, so be called the two-value gray level image.Maximum variance between clusters is a prior art, and it is a threshold value of finding out the maximum between-cluster variance maximum of cutting apart prospect (being target) and background according to gray value.As in the original chest image, remember that t is the segmentation threshold of prospect and background, prospect is counted and accounted for image scaled is w0, average gray is u0; Background is counted and accounted for image scaled is w1, and average gray is u1.The grand mean gray scale of image is: u=w0*u0+w1*u1.From the minimum gradation value to the maximum gradation value, travel through t, when t makes variance yields g=w0* (u0-u) 2+ w1* (u1-u) 2When maximum, t is the optimal threshold of cutting apart.Variance is the inhomogeneity a kind of tolerance of intensity profile, variance yields is big more, two parts difference that composing images is described is big more, is divided into target and all can causes two parts difference to diminish when part target mistake is divided into background or part background mistake, therefore makes to mean the misclassification probability minimum cutting apart of inter-class variance maximum.Lung wherein, pneumothorax, the lower parts of CT value such as trachea blood vessel are split together.
Step 22, remove the part of non-thorax among the original thorax figure of gray level threshold segmentation that from step 21, obtains, in the image that step 21 obtains, find maximum UNICOM zone and its border, remove all targets outside this border, obtain thorax, and chest is filled, obtain complete thorax image, shown in Fig. 3 c.Because can have very little, highdensity blood vessel in the pulmonary in thorax,, need to fill these little blood vessels with its part as thorax in order to obtain complete thorax image.
In order independently to calculate the size of the pneumothorax in the thorax of the left and right sides, need to calculate separately the area of left and right sides pleurobranch.Under pulmonary's normal condition, mainly have only lung in the thoracic cavity, pulmonary takes whole thoracic cavity.Because the front and rear edge of lung is very thin usually, so [adhesion can occur at front and rear edge sometimes, makes that left and right sides thorax connects together in the image, need separate left and right sides thorax, so that obtain isolating left and right sides thorax.By the thorax image is judged, whether need to carry out left and right sides thorax and separate, if carry out step 23; If obtain that left and right sides thorax has separated in the image, do not need to separate, then carry out step 24.
Judge that whether left and right sides thorax separates is to judge according to the number of connected region with according to the area of connected region whether left and right sides thorax connects together.Such as the zone of two connections is arranged, and the area in the zone of two connections to reach default value be 100 as the area of adult's one-sided thorax, can think that left and right sides thorax is isolating; If have only the zone of a connection, this explanation left and right sides thorax is to connect together, and need separate.Certainly, if though the zone of two connections is arranged, the area in a zone is very little, and an other area is very big, then illustrates what left and right sides thorax still connected together, and little connected region is a trachea, still needs left and right sides thorax is separated.
Step 23 is carried out the distance conversion to the thorax image that obtains in the step 22 and is obtained the thorax distance map, and carries out left and right sides thorax based on this thorax distance map and separate, and obtains isolating left and right sides thorax image, as Fig. 3 e.
As shown in Figure 4, shown left and right sides thorax separation steps:
Step 231, the two-value gray level image of the complete thorax image in the obtaining step 22;
Step 232 is called the expansion function earlier to the gray level image in the step 231, calls the corrosion function again and realizes closed operation, reaches level and smooth purpose;
Step 233 is carried out the conversion of gray scale and distance to the image in the step 232, produces a range image, shown in Fig. 3 d; The gray value of the every bit in the distance map is this point minimum range to background (black region) in the two-value gray-scale map.
Step 234 is provided with a high gray threshold, cuts apart range image, judges whether left and right sides thorax separates, if do not separate, resets a higher gray threshold, cuts apart range image, separates up to left and right sides thorax, obtains the distance map of isolating left and right sides thorax.
Step 235 reduces gray threshold gradually, and with the distance map of the isolating left and right sides thorax of acquisition in the segmentation procedure 234, thorax links up up to the left and right sides, and preserves result of this time cutting apart and the result of cutting apart last time;
Step 236 deducts the result of cutting apart last time by this result of cutting apart, obtains difference results;
Step 237 in segmentation result last time (cutting apart previous segmentation result for the last time), from top to bottom, from left to right travels through each and belongs to the interior point of difference results, detects its adjacent area;
Step 238 is cut apart the last time and is found out those points in the previous segmentation result, and its adjacent area comprises the point that is positioned at left thorax and right thorax simultaneously, these tie-points as left and right sides thorax that record is found;
Step 239 removes the junction point and these junction points that makes progress in front and back that are found in the thorax gray level image, separate up to left and right sides thorax, obtains the isolating image of left and right sides thorax, shown in Fig. 3 e.
Step 24 shifts out trachea, bronchus from the image of isolating left and right sides thorax, obtain an image as Fig. 3 f.
Since trachea, bronchus have the density that is similar to pneumothorax, they might be considered to pneumothorax, and the result who makes the mistake.So, these tracheas, bronchus are removed in order to obtain result accurately.Judge whether it is trachea or bronchus, can confirm from area and circularity, for those areas less than the gross area 2 ‰, circularity will be considered to trachea or bronchus greater than 0.9 and should remove, shown in Fig. 3 f.Sometimes, those circularities may be retained less than 0.9 trachea or bronchus, and the user can remove these trachea or bronchus by manual mode.Only removing according to area, is unacceptable, because some pneumothorax also is very little.
Can calculate by following mode for circularity:
RoundMetric = π × EquivDiameter Perimeter Formula 1
Wherein, EquivDiameter is an equivalent diameter, and it is the circular cross-section diameter that converses according to area-of-interest area equal principle;
Perimeter is meant the girth of area-of-interest.
The doctor can make amendment for the image that obtains in the step 23,24, such as by gray threshold, area threshold, circularity threshold value, cut apart automatically with the mode of manual modification image and make amendment again.
Wherein in the step 3, cut apart pneumothorax by following mode:
Step 31, the thorax position, the left and right sides that splits according to step 2 extracts the original image of thorax from original chest image, and shown in Fig. 5 a, to simplify follow-up segmentation procedure, this image is many gray-value image.
Step 32 is carried out threshold value to the original thorax image that extracts in the step 31 and is cut apart, and obtains to have only the image of pneumothorax, shown in Fig. 5 b; Gray value and CT value have corresponding relation in the CT image, can be the CT value also so this threshold value can be a gray value, and what adopt among the present invention is that the CT value is cut apart.Because in the pneumothorax mainly is that air is arranged, and cuts apart so can carry out threshold value based on the CT value of air.This CT value can be a default.As the CT value of air is about-1000HU, then can set a CT threshold value to be-900HU about, as long as be equal to or less than this threshold value, be exactly air.Air CT value threshold value by setting can split pneumothorax.
For in the step 4, calculate the ratio of pneumothorax in thorax, comprise following step:
Step 41, classification about the pneumothorax result of cutting apart carried out;
Because the ratio of calculating pneumothorax is according to calculatings of coming of left and right sides thorax branch, so classify about the result in step 2 and the step 3 carried out according to geometrical property such as barycenter;
Step 42, the ratio of pneumothorax in the thorax of the calculating left and right sides;
Calculate the ratio of pneumothorax in the thorax of the left and right sides according to following formula 2:
PTXpercentage = Σ i = 1 N PTXarea Σ i = 1 N HemithoraxArea Formula 2
Wherein, Hemithorax Area represents the left thorax in each cross section or the area of right thorax, comprises left lung or right lung tissue that pneumothorax that each cross section splits adds each cross section;
PTX area represents the area of the pneumothorax that each cross section splits;
The quantity in ' N ' expression cross section.
The X ray CT device of measurement pneumothorax of the present invention comprises: scan table 100, and it carries out ray scanning to person under inspection's chest; Data collection module 200, its scan-data to scan table output is collected, analog digital conversion; Image reconstruction unit 300, its scan-data that sends according to data collection module carries out image reconstruction, and the image data storage that institute's reconstruct is obtained is to memory element 700; User interface 400 for user's operation; With central control unit 500, it is connected to the outfan of described image reconstruction unit 300, and gated sweep platform 100, data collection module 200 and user interface 400; Central control unit 500 receives the operational order that user interface 400 sends, and the image that image reconstruction unit is sent is handled, and shown in user interface 400; With pneumothorax measuring unit 600, its chest image after to image reconstruction unit 300 reconstruct carries out pneumothorax and measures.
Described pneumothorax measuring unit 600 comprises left and right sides thorax cutting unit 601, pneumothorax cutting unit 602, computing unit 603.Wherein, left and right sides thorax cutting unit 601 carries out left and right sides thorax to original chest image and cuts apart under the control of central control unit 500, obtains the isolating thorax image of left and right sides thorax.602 pairs of original chest images of described pneumothorax cutting unit are cut apart, and obtain to have only the image of pneumothorax.Computing unit 603 is cut apart the area of the isolating thorax image calculation of the left and right sides thorax left and right sides thorax of acquisition according to left and right sides thorax cutting unit 601; Cut apart the pneumothorax image calculation pneumothorax area of acquisition according to pneumothorax cutting unit 602, and calculate the ratio of pneumothorax according to formula (2).
Wherein, described left and right sides thorax cutting unit 601 also comprises thorax acquiring unit 604, judging unit 605 and left and right sides thorax separative element 606; Wherein 604 pairs of original chest images of thorax acquiring unit (shown in Fig. 3 a) carry out gray level threshold segmentation, obtain threshold value and cut apart the original thorax image in back, this image is the two-value gray level image, shown in Fig. 3 b, this image has only B﹠W, so be called the two-value gray level image.This thresholding method can adopt maximum variance between clusters, as mentioned above, is not repeated at this.Wherein the lower parts of CT value such as lung, pneumothorax, trachea, blood vessel are split together.From the two-value gray level image that obtains, remove the part of non-thorax such as rib, bed, find maximum UNICOM zone and its border, remove all targets outside this border, obtain the thorax zone.And the thorax zone filled, thereby obtain complete thorax image, shown in Fig. 3 c.Because in intrathoracic pulmonary, can have very little, highdensity blood vessel,, need to fill these little blood vessels with its part as thorax in order to obtain complete thorax image.
The front and rear edge of lung is very thin usually, so [adhesion can occur at front and rear edge sometimes, makes left and right sides thorax connect together, and in order independently to calculate the size of the pneumothorax in the thorax of the left and right sides, need separate left and right sides thorax.The thorax image that 605 pairs of thorax acquiring units 604 of judging unit obtain judges judge whether left and right sides thorax separates, and instruction is sent to central control unit 500, and central control unit 500 is according to instruction, the work of control left and right sides thorax separative element.If it is isolating that judged result needs, 606 pairs of thorax images of central control unit 500 control left and right sides thorax separative elements are cut apart, and separate left and right sides breast exterior feature; If judged result is not need isolatingly, then controls left and right sides thorax separative element 606 and do not work.
Judging unit 605 judges according to the number of connected region with according to the area of connected region whether left and right sides thorax connects together.Such as the zone of two connections is arranged, and the area in the zone of two connections to reach default value be 100 as the area of adult's one-sided thorax, can think that left and right sides thorax is isolating; If have only the zone of a connection, this explanation [is to connect together, and need separate left and right sides thorax.Certainly, if though the zone of two connections is arranged, the area in a zone is very little, and an other area is very big, then illustrates what [still connected together, and little connected region is a trachea, still needs left and right sides thorax is separated.
The needs that thorax separative element 606 pairs of thorax acquiring units 604 in the left and right sides send carry out isolating thorax image and carry out distance conversion acquisition distance map, and carry out left and right sides thorax based on this distance map and separate, and obtain isolating left and right sides thorax image, as Fig. 3 e.Left and right sides thorax separative element 606 carries out left and right sides thorax by aforesaid step 231 to the mode of step 239 to be separated, and obtains the isolating thorax contour images of left and right sides thorax.
Pneumothorax cutting unit 602 extracts original thorax image from original chest image, shown in Fig. 5 a.This extracting method can be the image that extracts original thorax according to position correspondence from original chest image of the left and right sides thorax image that obtains in the thorax acquiring unit 604.Afterwards the image that extracts is carried out threshold value and cut apart, obtain to have only the image of pneumothorax, shown in Fig. 5 b; This threshold value can be that gray value can be the CT value also, and what adopt among the present invention is that the CT value is cut apart.Because in the pneumothorax mainly is air, cuts apart so can carry out threshold value based on the CT value of air.This CT value can be a default.As the CT value of air is about-1000HU, then can set a CT threshold value to be-900HU about, as long as be equal to or less than this threshold value, be exactly air.Air CT value threshold value by setting can split pneumothorax.
Computing unit 603 also comprises taxon 607, classification about its pneumothorax result that pneumothorax cutting unit 60 is cut apart carries out; Because the ratio of calculating pneumothorax is according to calculatings of coming of left and right sides thorax branch, so classify about need carrying out according to geometrical property such as barycenter the pneumothorax segmentation result.
As shown in Figure 7, described user interface 4 comprises display interface and operation interface, display interface is used for display image such as original chest image, the image in each stage comprised two-value gray level image, thorax image, thorax distance map, isolating left and right sides thorax image etc. during left and right sides thorax separated, and the image in the pneumothorax cutting procedure.Operation interface comprises that automatic pneumothorax computing function button, thorax are cut apart button, pneumothorax is cut apart button, pneumothorax computed push-buttom; Wherein cut apart button, pneumothorax and cut apart the corresponding respectively input frame that is provided with gray scale, area, circular arc degree under the button at thorax.The user can select automatic pneumothorax computing function button that pneumothorax is calculated automatically, measures; Operation thorax that also can be manual is cut apart button, pneumothorax is cut apart button, pneumothorax computed push-buttom, revises the interior value of input frame of gray scale, area, circular arc degree, carries out manual operation, calculates pneumothorax.

Claims (21)

1. X ray CT device of measuring pneumothorax comprises:
Scan table, it carries out ray scanning to person under inspection's chest;
Data collection module, its scan-data to scan table output is collected, analog digital conversion;
Image reconstruction unit, its scan-data that sends according to data collection module carries out image reconstruction, and the image data storage that institute's reconstruct is obtained is to a memory element;
User interface for user's operation; With
Central control unit, it is connected to the outfan of described image reconstruction unit, and gated sweep platform, data collection module and user interface; Central control unit receives the operational order that user interface sends, and the image that image reconstruction unit is sent is handled, and shown in user interface; It is characterized in that, also comprise:
Pneumothorax measuring unit, its chest image after to image reconstruction unit reconstruct carry out pneumothorax and measure.
2. the X ray CT device of measurement pneumothorax as claimed in claim 1 is characterized in that: described pneumothorax measuring unit comprises left and right sides thorax cutting unit, pneumothorax cutting unit, computing unit; Wherein, left and right sides thorax cutting unit carries out left and right sides thorax to original chest image and cuts apart under the control of central control unit, obtains the isolating thorax image of left and right sides thorax; The position of the left and right sides thorax image that described pneumothorax cutting unit splits according to left and right sides thorax cutting unit extracts original thorax image from original chest image, and this original thorax image is cut apart, and obtains the pneumothorax image; Computing unit is cut apart the area of the isolating thorax image calculation of the left and right sides thorax left and right sides thorax of acquisition according to left and right sides thorax cutting unit; Cut apart the pneumothorax image calculation pneumothorax area of acquisition according to the pneumothorax cutting unit, and calculate the ratio of pneumothorax in the thorax of the left and right sides according to following formula,
PTXpercentage = Σ i = 1 N PTXarea Σ i = 1 N HemithoraxArea
Wherein, Hemithorax Area represents the left thorax in each cross section or the area of right thorax, comprises left lung or right lung tissue that pneumothorax that each cross section splits adds each cross section;
PTX area represents the area of the pneumothorax that each cross section splits;
The quantity in ' N ' expression cross section.
3. the X ray CT device of measurement pneumothorax as claimed in claim 2 is characterized in that:
Described left and right sides thorax cutting unit also comprises thorax acquiring unit, judging unit and left and right sides thorax separative element; Wherein the thorax acquiring unit carries out gray level threshold segmentation to original chest image, obtains threshold value and cuts apart the original thorax image in back, and this image is the two-value gray level image; Judging unit judges the thorax image that the thorax acquiring unit obtains judge whether left and right sides thorax separates, and instruction is sent to central control unit, and central control unit is according to instruction, the work of control left and right sides thorax separative element; If it is isolating that judged result needs, central control unit control left and right sides thorax separative element is cut apart the thorax image, separates left and right sides breast exterior feature; If judged result is not need isolatingly, then controls left and right sides thorax separative element and do not work; The needs that left and right sides thorax separative element sends the thorax acquiring unit carry out isolating thorax image and carry out left and right sides thorax separation.
4. the X ray CT device of measurement pneumothorax as claimed in claim 3 is characterized in that:
Wherein original chest image is carried out gray level threshold segmentation is to adopt maximum variance between clusters to the thorax acquiring unit; The thorax acquiring unit removes the part of non-thorax from the two-value gray level image that obtains, and finds maximum UNICOM zone and its border, removes all targets outside this border, obtains the thorax zone; And the thorax zone filled, obtain complete thorax image.
5. the X ray CT device of measurement pneumothorax as claimed in claim 4 is characterized in that:
Wherein judging unit judges according to the number of connected region with according to the area of connected region whether left and right sides thorax connects together.
6. the X ray CT device of measurement pneumothorax as claimed in claim 5 is characterized in that:
Wherein left and right sides thorax separative element needs that the thorax acquiring unit is sent carry out isolating thorax image and carry out the distance conversion and obtain distance map, and carry out left and right sides thorax based on this distance map and separate, obtain isolating left and right sides thorax image, and from the image of isolating left and right sides thorax, shift out trachea, bronchus.
7. the X ray CT device of measurement pneumothorax as claimed in claim 6 is characterized in that:
Left and right sides thorax separative element carries out left and right sides thorax by following step to be separated, and obtains the isolating thorax contour images of left and right sides thorax:
Step 231 is obtained the two-value gray level image of complete thorax image;
Step 232 is called the expansion function earlier to the gray level image in the step 231, calls the corrosion function again and realizes closed operation, reaches level and smooth purpose;
Step 233 is carried out the conversion of gray scale and distance to the image in the step 232, produces a range image; The gray value of the every bit in the distance map is this point minimum range to background in the two-value gray-scale map;
Step 234 is provided with a high gray threshold, cuts apart range image, judges whether left and right sides thorax separates, if do not separate, resets a higher gray threshold, cuts apart range image, separates up to left and right sides thorax, obtains the distance map of isolating left and right sides thorax;
Step 235 reduces gray threshold gradually, and with the distance map of the isolating left and right sides thorax of acquisition in the segmentation procedure 234, thorax links up up to the left and right sides, and preserves result of this time cutting apart and the result of cutting apart last time;
Step 236 deducts the result of cutting apart last time by this result of cutting apart, obtains difference results;
Step 237 in the last time segmentation result, from top to bottom, from left to right travels through each and belongs to point in the difference results, detects its adjacent area;
Step 238 is cut apart the last time and is found out those points in the previous segmentation result, and its adjacent area comprises the point that is positioned at left thorax and right thorax simultaneously, these tie-points as left and right sides thorax that record is found;
Step 239 removes the junction point and these junction points that makes progress in front and back that are found in the thorax gray level image, separate up to left and right sides thorax, obtains the isolating image of left and right sides thorax.
8. the X ray CT device of measurement pneumothorax as claimed in claim 6, it is characterized in that: confirm trachea, bronchus according to region area and circularity, for area less than the gross area 2 ‰, circularity is considered to trachea or bronchus greater than 0.9 and is removed; Circularity can be calculated by following mode:
RoundMetric = π × EquivDiameter Perimeter
Wherein, EquivDiameter is an equivalent diameter, and it is the circular cross-section diameter that converses according to area-of-interest area equal principle; Perimeter is meant the girth of area-of-interest.
9. the X ray CT device of measurement pneumothorax as claimed in claim 2 is characterized in that:
The pneumothorax cutting unit carries out threshold value to the original thorax image that extracts to be cut apart, and obtains to have only the image of pneumothorax.
10. the X ray CT device of measurement pneumothorax as claimed in claim 9 is characterized in that: original thorax image is carried out CT value threshold value cut apart, obtain the pneumothorax image.
11. the X ray CT device of measurement pneumothorax as claimed in claim 10 is characterized in that: computing unit also comprises taxon, classification about it carries out the pneumothorax result according to geometrical property.
12. X ray CT device as claim 7 or 11 described measurement pneumothoraxs, it is characterized in that: described user interface comprises display interface and operation interface, display interface is used to show original chest image, the image in each stage comprised two-value gray level image, thorax image, thorax distance map, isolating left and right sides thorax image during left and right sides thorax separated, and the image in the pneumothorax cutting procedure.
13. the X ray CT device of measurement pneumothorax as claimed in claim 12 is characterized in that: operation interface comprises that automatic pneumothorax computing function button, thorax are cut apart button, pneumothorax is cut apart button, pneumothorax computed push-buttom; Wherein cut apart button, pneumothorax and cut apart the corresponding respectively input frame that is provided with gray scale, area, circular arc degree under the button at thorax; The user can select automatic pneumothorax computing function button that pneumothorax is calculated automatically, measures; Operation thorax that also can be manual is cut apart button, pneumothorax is cut apart button, pneumothorax computed push-buttom, revises the interior value of input frame of gray scale, area, circular arc degree, carries out manual operation, calculates pneumothorax.
14. a method of measuring pneumothorax comprises the steps:
Step 1 is carried out CT scan to chest, obtains all original chest images;
Step 2 is cut apart the thoracic cavity each original chest image, obtains the complete isolating image of left and right sides thorax;
Step 3 is cut apart pneumothorax each original chest image, isolates pneumothorax, obtains the image of pneumothorax;
Step 4 is calculated the pneumothorax ratio in every one-sided thorax.
15. the method for measurement pneumothorax as claimed in claim 14 is characterized in that: obtain the isolating image of left and right sides thorax in the step 2 and comprise following step:
Step 21 is carried out automatic gray level threshold segmentation based on maximum variance between clusters to original chest image, obtains threshold value and cuts apart the original thorax figure in back, and this original thorax figure is the two-value gray level image;
Step 22 removes the part of non-thorax among the original thorax figure of gray level threshold segmentation that obtains from step 21, obtain complete thorax image;
By the thorax image is judged, whether need to carry out left and right sides thorax and separate, if carry out step 23; If obtain that left and right sides thorax has separated in the image, do not need to separate, then carry out step 24;
Step 23 is carried out the distance conversion to the thorax image that obtains in the step 22 and is obtained the thorax distance map, and carries out left and right sides thorax based on this thorax distance map and separate, and obtains isolating left and right sides thorax image;
Step 24 shifts out trachea, bronchus from the image of isolating left and right sides thorax.
16. the method for measurement pneumothorax as claimed in claim 15, it is characterized in that: in the image that step 21 obtains, find maximum UNICOM zone and its border, remove all targets outside this border, obtain thorax, and chest is filled, obtain complete thorax image, fill very little, highdensity blood vessel in the pulmonary in the thorax, obtain complete thorax image.
17. the method for measurement pneumothorax as claimed in claim 15 is characterized in that, step 23 is carried out left and right sides thorax by following step and is separated:
Step 231, the two-value gray level image of the complete thorax image in the obtaining step 22;
Step 232 is called the expansion function earlier to the gray level image in the step 231, calls the corrosion function again and realizes closed operation, reaches level and smooth purpose;
Step 233 is carried out the conversion of gray scale and distance to the image in the step 232, produces a range image; The gray value of the every bit in the distance map is this point minimum range to background in the two-value gray-scale map;
Step 234 is provided with a high gray threshold, cuts apart range image, judges whether left and right sides thorax separates, if do not separate, resets a higher gray threshold, cuts apart range image, separates up to left and right sides thorax, obtains the distance map of isolating left and right sides thorax;
Step 235 reduces gray threshold gradually, and with the distance map of the isolating left and right sides thorax of acquisition in the segmentation procedure 234, thorax links up up to the left and right sides, and preserves result of this time cutting apart and the result of cutting apart last time;
Step 236 deducts the result of cutting apart last time by this result of cutting apart, obtains difference results;
Step 237 in the last time segmentation result, from top to bottom, from left to right travels through each and belongs to point in the difference results, detects its adjacent area;
Step 238 is cut apart the last time and is found out those points in the previous segmentation result, and its adjacent area comprises the point that is positioned at left thorax and right thorax simultaneously, these tie-points as left and right sides thorax that record is found;
Step 239 removes the junction point and these junction points that makes progress in front and back that are found in the thorax gray level image, separate up to left and right sides thorax, obtains the isolating image of left and right sides thorax.
18. the method for measurement pneumothorax as claimed in claim 16, it is characterized in that: judge whether it is trachea or bronchus by region area and circularity, for those areas less than the gross area 2 ‰, circularity is considered to trachea or bronchus greater than 0.9 and should removes; Circularity is calculated by following mode:
RoundMetric = π × EquivDiameter Perimeter
Wherein, EquivDiameter is an equivalent diameter, and it is the circular cross-section diameter that converses according to area-of-interest area equal principle; Perimeter is meant the girth of area-of-interest.
19. the method for measurement pneumothorax as claimed in claim 14 is characterized in that: wherein in the step 3, cut apart pneumothorax by following mode:
Step 31, according to the position of the left and right sides thorax image that obtains in the step 2 from the original thorax image of extracting of original chest image correspondence;
Step 32 is carried out threshold value to the original thorax image that extracts in the step 31 and is cut apart, and obtains to have only the image of pneumothorax.
20. the method for measurement pneumothorax as claimed in claim 19 is characterized in that: adopt the CT value to cut apart to original thorax image, obtain to have only the image of pneumothorax.
21. the method for measurement pneumothorax as claimed in claim 19 is characterized in that, in the step 4, calculates the ratio of pneumothorax in thorax, comprises following step:
Step 41, classification about according to geometrical property the pneumothorax result of cutting apart being carried out;
Step 42 is according to the ratio of pneumothorax in the following formula calculating left and right sides thorax;
PTXpercentage = Σ i = 1 N PTXarea Σ i = 1 N HemithoraxArea
Wherein, Hemithorax Area represents the left thorax in each cross section or the area of right thorax, comprises left lung or right lung tissue that pneumothorax that each cross section splits adds each cross section; PTX area represents the area of the pneumothorax that each cross section splits;
The quantity in ' N ' expression cross section.
CN201010180333.9A 2010-05-14 2010-05-14 Method and apparatus for measuring pneumothorax Active CN102240212B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201010180333.9A CN102240212B (en) 2010-05-14 2010-05-14 Method and apparatus for measuring pneumothorax

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201010180333.9A CN102240212B (en) 2010-05-14 2010-05-14 Method and apparatus for measuring pneumothorax

Publications (2)

Publication Number Publication Date
CN102240212A true CN102240212A (en) 2011-11-16
CN102240212B CN102240212B (en) 2015-03-25

Family

ID=44958588

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201010180333.9A Active CN102240212B (en) 2010-05-14 2010-05-14 Method and apparatus for measuring pneumothorax

Country Status (1)

Country Link
CN (1) CN102240212B (en)

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107665497A (en) * 2016-07-29 2018-02-06 上海联影医疗科技有限公司 In a kind of medical image calculate ambition than method
WO2018205922A1 (en) * 2017-05-08 2018-11-15 Suzhou Complexis Medical Inc. Methods and systems for pulmonary function test based on diagnostic medical imaging and machine learning
CN109816672A (en) * 2019-02-25 2019-05-28 数坤(北京)网络科技有限公司 A kind of image segmentation acquisition methods and device for neck body bone
CN109919935A (en) * 2019-03-12 2019-06-21 语坤(北京)网络科技有限公司 A kind of neck body blood vessel segmentation method and apparatus
CN111415341A (en) * 2020-03-17 2020-07-14 北京推想科技有限公司 Pneumonia stage evaluation method, pneumonia stage evaluation device, pneumonia stage evaluation medium and electronic equipment
CN111449669A (en) * 2014-07-02 2020-07-28 柯惠有限合伙公司 System and method for detecting trachea
CN112712508A (en) * 2020-12-31 2021-04-27 杭州依图医疗技术有限公司 Method and device for determining pneumothorax
CN114332023A (en) * 2021-12-30 2022-04-12 上海市嘉定区中心医院 Pneumothorax automatic diagnosis and crisis early warning method, device, equipment and storage medium

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
SU426655A1 (en) * 1971-02-02 1974-05-05 П. П. Герцен Оренбургский сельскохоз йственный институт METHOD FOR CARRYING OUT OPERATIONS ON BODIES
SU724130A1 (en) * 1978-03-20 1980-03-30 Днепропетровский Ордена Трудового Красного Знамени Медицинский Институт Closed pneumothorax treatment method
FR2538705A1 (en) * 1982-12-29 1984-07-06 Vincent Michel MEDICAL APPARATUS FOR EXTENDING PNEUMOTHORAX
US5668888A (en) * 1990-11-21 1997-09-16 Arch Development Corporation Method and system for automatic detection of ribs and pneumothorax in digital chest radiographs
CN1512853A (en) * 2001-06-01 2004-07-14 ��ʽ��������ҽҩ X-ray CT device, image processor, and method for processing imatge of X-ray CT device
CN1537513A (en) * 2003-04-18 2004-10-20 ��ʽ���綫֥ X-ray computed tomography apparatus

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
SU426655A1 (en) * 1971-02-02 1974-05-05 П. П. Герцен Оренбургский сельскохоз йственный институт METHOD FOR CARRYING OUT OPERATIONS ON BODIES
SU724130A1 (en) * 1978-03-20 1980-03-30 Днепропетровский Ордена Трудового Красного Знамени Медицинский Институт Closed pneumothorax treatment method
FR2538705A1 (en) * 1982-12-29 1984-07-06 Vincent Michel MEDICAL APPARATUS FOR EXTENDING PNEUMOTHORAX
US5668888A (en) * 1990-11-21 1997-09-16 Arch Development Corporation Method and system for automatic detection of ribs and pneumothorax in digital chest radiographs
JP3134009B2 (en) * 1990-11-21 2001-02-13 アーチ・デベロップメント・コーポレーション Image processing method and apparatus
CN1512853A (en) * 2001-06-01 2004-07-14 ��ʽ��������ҽҩ X-ray CT device, image processor, and method for processing imatge of X-ray CT device
CN1537513A (en) * 2003-04-18 2004-10-20 ��ʽ���綫֥ X-ray computed tomography apparatus

Non-Patent Citations (5)

* Cited by examiner, † Cited by third party
Title
JEAN M.SEELY,ET AL.,: "High-Resolution CT of Pediatric Lung Disease:Imaging Findings", 《AJR》 *
SHIGERU SANADA,ET AL.,: "Image feature analysis and computer-aided diagnosis in digital radiography:Automated detection of pneumothorax in chest images", 《MEDICAL PHYSICS》 *
梁建辉等: "CT测量气胸肺压缩程度的动物实验研究", 《国际医药卫生导报》 *
王元文等: "自发性气胸肺压缩程度的计算方法与评价", 《自发性气胸肺压缩程度的计算方法和评价》 *
王成林等: "气胸肺压缩率的CT测量诊断", 《中华创伤杂志》 *

Cited By (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111449669A (en) * 2014-07-02 2020-07-28 柯惠有限合伙公司 System and method for detecting trachea
CN111449669B (en) * 2014-07-02 2023-04-18 柯惠有限合伙公司 System and method for detecting trachea
CN107665497A (en) * 2016-07-29 2018-02-06 上海联影医疗科技有限公司 In a kind of medical image calculate ambition than method
CN107665497B (en) * 2016-07-29 2020-11-20 上海联影医疗科技有限公司 Method for calculating cardiothoracic ratio in medical image
WO2018205922A1 (en) * 2017-05-08 2018-11-15 Suzhou Complexis Medical Inc. Methods and systems for pulmonary function test based on diagnostic medical imaging and machine learning
CN109816672A (en) * 2019-02-25 2019-05-28 数坤(北京)网络科技有限公司 A kind of image segmentation acquisition methods and device for neck body bone
CN109816672B (en) * 2019-02-25 2021-02-02 语坤(北京)网络科技有限公司 Image segmentation acquisition method and device for head and neck bones
CN109919935A (en) * 2019-03-12 2019-06-21 语坤(北京)网络科技有限公司 A kind of neck body blood vessel segmentation method and apparatus
CN109919935B (en) * 2019-03-12 2021-02-09 语坤(北京)网络科技有限公司 Head and neck blood vessel segmentation method and device
CN111415341A (en) * 2020-03-17 2020-07-14 北京推想科技有限公司 Pneumonia stage evaluation method, pneumonia stage evaluation device, pneumonia stage evaluation medium and electronic equipment
CN112712508A (en) * 2020-12-31 2021-04-27 杭州依图医疗技术有限公司 Method and device for determining pneumothorax
CN114332023A (en) * 2021-12-30 2022-04-12 上海市嘉定区中心医院 Pneumothorax automatic diagnosis and crisis early warning method, device, equipment and storage medium

Also Published As

Publication number Publication date
CN102240212B (en) 2015-03-25

Similar Documents

Publication Publication Date Title
CN102240212B (en) Method and apparatus for measuring pneumothorax
CN110796613B (en) Automatic identification method and device for image artifacts
CN107622492B (en) Lung fissure segmentation method and system
JP4728627B2 (en) Method and apparatus for segmenting structures in CT angiography
US7295870B2 (en) Method for the detection and automatic characterization of nodules in a tomographic image and a system of medical imaging by tomodensimetry
US7813536B2 (en) Image measuring apparatus and method, and image measuring system for glomerular filtration rate
CN106447645A (en) Device and method for coronary artery calcification detection and quantification in CTA image
WO2012112627A4 (en) Method and apparatus of cone beam breast ct image-based computer-aided detection and diagnosis
CN110503635B (en) Hand bone X-ray film bone age assessment method based on heterogeneous data fusion network
WO2005009240A9 (en) Image processing method and computer-readable recording medium containing image processing program
CN110796670A (en) Dissection method and device for dissecting artery
JP2000126168A (en) Bone image processing method and bone strength evaluation method
CN113222951B (en) Osteoporosis artificial intelligent diagnosis device for recognizing hip joint X-ray
CN107708550A (en) For the surface modeling for the segmentation acoustic echo structure for detecting and measuring anatomic abnormalities
CN114261095B (en) AI-based orthopedic 3D printing method and device
KR101105494B1 (en) A reconstruction method of patient-customized 3-D human bone model
CN110874860A (en) Target extraction method of symmetric supervision model based on mixed loss function
CN111325754B (en) Automatic lumbar vertebra positioning method based on CT sequence image
US9275452B2 (en) Method and system for automatically determining compliance of cross sectional imaging scans with a predetermined protocol
CN114972376A (en) Coronary calcified plaque segmentation method, segmentation model training method and related device
EP2372643A2 (en) System and method of pulmonary emboli detection for computed tomography
CN110738633A (en) organism tissue three-dimensional image processing method and related equipment
US20040105527A1 (en) Methods and apparatus for the classification of nodules
EP3501399A1 (en) Method of quantification of visceral fat mass
CN114742753A (en) Image evaluation method and device based on neural network

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant